Hope1337

[Bug] Inference video not showed

1

### Branch main branch (1.x version, such as `v1.0.0`, or `dev-1.x` branch) ### Prerequisite - [X] I have searched [Issues](https://github.com/open-mmlab/mmaction2/issues) and [Discussions](https://github.com/open-mmlab/mmaction2/discussions) but cannot get the expected help. - [X]...

In augmetation file, at function `RandomSampleCrop`, we use np.random.choice, but I guess that there was some change in numpy lib or python so we can't use it for tupe like...

performence with not clipped prior boxes

1

I mentioned in comment #94 that the prior boxes were not clipped correctly. I intended to compare the performance of using correct prior boxes versus "wrong" prior boxes, and just...

prior box not clipped as expected

12

 As you stated, prior box will be clipped if it out side of image, but in your code : ```python prior_boxes.clamp_(0, 1) # (8732, 4) ``` prior_boxes still in...

How can train using my own dataset

2

Same as title : How can train using my own dataset ?

Chào anh Tiệp, em là người đọc sách của anh, trong quá trình đọc em đã phát hiện ra một số lỗi, topic này chỉ để đóng góp chứ không...

Difference between original paper and code

1

In original paper (on page 2, section 2.2) the authors said that they keep last 100 feature for each track to calculate the cosine distance metric :  But in...

Initialization of yolov8

1

### Search before asking - [X] I have searched the YOLOv8 [issues](https://github.com/ultralytics/ultralytics/issues) and [discussions](https://github.com/ultralytics/ultralytics/discussions) and found no similar questions. ### Question I found that YOLOv8 does not use initialization weights....

Hi, thanks for great works. I found that [YOLOv8 use C2f module](https://github.com/ultralytics/ultralytics/issues/189) instead of CSP module as in your code. Is that a mistake or intentional on your part?

] In the numerator, I think it should be $p(y|x, \theta)p(x|\theta)p(\theta)$, so why $p(x|\theta)$ disappear ?